Kinematic design optimization for anthropomorphic robot hand based on interactivity of fingers

The human hand provides excellent grasp capabilities by using the interaction of the thumb and other fingers, and thus, how to dextrously interact with each other can represent the performance of the robot hand. This paper presents the kinematic design optimization for an anthropomorphic robot hand based on the interactivity of fingers. We propose a new performance index, called ‘interactivity of fingers (IF),’ which is useful for quantifying the multifingered precision grasping capability of the robot hand. Using the IF, the kinematic model of an anthropomorphic robot hand is optimized. The optimization is performed by using the genetic algorithm, and position and orientation of the thumb’s saddle joint are systematically determined. In order to verify the usefulness of the proposed performance index, IF of existing hands and that of the optimized model are compared. As a result, the IF of the optimized robot hand is almost five times larger than those of the other hands and the effectiveness of the proposed method is discussed.